Endoscopic snare resection followed by laser ablation in the treatment of large,

Transcription

1 Scandinavian Journal of Surgery 00: 99 04, 20 Endoscopic snare resection followed by laser ablation in the treatment of large, sessile rectal adenomas A. Nesbakken, 2, J. Kristinsson, A. Svindland 2, 3, O. C. Lunde Department of Gastrointestinal Surgery, Oslo University Hospital, Aker, Oslo, Norway 2 Faculty of Medicine, University of Oslo, Oslo, Norway 3 Department of Pathology, Oslo University Hospital, Aker, Oslo, Norway Abstract Background and Aims: Large sessile rectal adenomas can be difficult to eradicate, and different treatment modalities are available. The aim of this study was to evaluate outcome after endoscopic snare resection followed by Nd:YAG laser ablation. Material and Methods: Over a 0-year period 92 of 99 (93%) patients were registered prospectively and attended follow-up examinations with endoscopy and biopsies. Results: Fifty-four (59%) men and 38 (4%) women were included; 67 patients (73%) had high grade (severe) intraepithelial dysplasia or intramucosal neoplasia. The adenomas ranged from 2 9 cm (median 4 cm) in diameter, and were located 2 5 cm (median 5 cm) from the anal verge. A median of two (range 6) piecemeal snare resection sessions and a median of one (range 7) laser treatments were performed for each patient. Complete eradication was achieved in 86 patients (93%). Over a median follow-up period of 26 months, 20/86 (23%) suffered local recurrence, eight of whom were given a second laser treatment without developing further recurrence. In five of eight frail patients considered unsuitable for more radical treatment, repeated laser treatment was effective in keeping the adenoma small and symptoms at a minimum. As a whole the treatment was successful in 74/92 (80%) and partially successful in 5/92 (5%) of the patients. Conclusions: Snare resection followed by laser ablation is safe and still has a place in the treatment of old, frail patients with large rectal adenomas. However, there is a risk of missing an infiltrating carcinoma, and other treatment options are preferable in fit patients. Key words: Adenoma; rectal neoplasms; endoscopy; lasers; laser therapy; endoscopic polypectomy; laser ablation; argon plasma coagulation; transanal endoscopic microsurgery (TEM) Correspondence: A. Nesbakken, M.D. Department of Gastrointestinal Surgery Oslo University Hospital, Aker NO Oslo Norway arild.nesbakken@aker-sykehus.no

2 00 A. Nesbakken, J. Kristinsson, A. Svindland, O. C. Lunde Introduction Treating large polyps in the rectum is challenging. Differentiating between in-situ and infiltrating carcinoma is difficult, even after thorough preoperative workout with endoscopic examination, digital rectal examination, transrectal ultrasound and biopsies. Several treatment options are available, including endoscopic procedures such as piecemeal snare resection ( 3), endoscopic mucosa resection (EMR)(4) and endoscopic submucosal dissection (ESD)(5), with or without supplementary ablation using methods such as argon plasma coagulation (APC), laser photocoagulation(6) and others. Surgical resection can be performed by conventional transanal excision (TAE)(7), transanal endoscopic microsurgery (TEM) (8) or formal rectal resection. The procedure of choice depends on the potential malignancy of the lesion, the exact location, especially the height above the dentate line, the patient s age and general condition, comorbidity and the available equipment and expertise. At our hospital the standard treatment for large rectal adenomas with no preoperative signs of infiltrating carcinoma used to be endoscopic snare resection followed by laser ablation. In the last few years this treatment has often been replaced by TEM in fit patients. TEM is a more radical, but also a more resource-demanding procedure. Since it is performed under general anaesthesia, and resection followed by bowel anastomosis is performed, there is a greater risk for the patient. The aim of this study was therefore to investigate outcome after snare resection followed by laser ablation, and evaluate the future role of this method of treatment. Material and methods Patients undergoing snare resection and laser treatment at our hospital were registered prospectively over a 0-year period. The details of the endoscopic examination, treatment sessions and follow-up examinations were recorded. Snare resection Flexible colonoscopes were used and snare resection was usually performed with a large monopolar snare and highfrequency diathermy current. Usually the lesions were so large that removal in one piece was not possible, and piecemeal resections were performed to remove as much as possible of visible neoplastic tissue. If necessary, snare resections were repeated every two weeks until this was achieved. All resected tissue was subjected to routine histopathological examination. As a part of the present study all slides were reviewed by a single, highly experienced pathologist (AS), to ensure correct classification of the adenoma. Laser ablation Four to six weeks after completing endoscopic resection, laser ablation was performed with a Neodynium-Yttrium- Aluminium garnet (Nd:YAG) laser (Sharplan 3000, Laser Industries Ltd, Tel Aviv, Israel). The fibre carrying the invisible laser light and the visible red helium-neon aiming beam was introduced through the working channel of a flexible colonoscope. The power was normally set at 40 watts and repeated energy pulses lasting 4 seconds were delivered. A continuous flow of coaxial CO 2 gas was used for cleaning and cooling the tip of the fibre. The procedure was usually performed under sedation, but spinal anaesthesia was used when the lesion was located close to the dentate line. The whole of the scarred area after snare resection, and any residual adenoma, was treated with the aim of vaporizing the tissue down to the border between the submucosa and the muscularis propria. As the Nd:YAG laser may penetrate too deeply, every effort was made to keep the laser beam parallel to the bowel wall in order to avoid perforation. Follow-up A follow-up endoscopy was performed within eight weeks, and if it revealed remnants of the adenoma, laser treatment usually without new snare resection was repeated until the tumour had been completely removed. The patients attended follow-up examinations with endoscopy every three months for a year, and thereafter every six months for a total of three years. Endoscopy and digital rectal examination were performed and biopsies were taken from the scar area and sites of suspected recurrence. Evaluation of treatment results When both endoscopy and histology revealed no residue of adenoma after the snare resection(s) and the first laser treatment(s), the result was classified as initially successful. When the tumour recurred after initial complete removal, this was classified as recurrence. When the adenoma had not been completely removed after repeated laser treatments, the treatment was classified as initial failure. If the patient was old and frail and not fit for more radical treatment after initial failure or recurrence, and if it was possible to keep the adenoma small by means of repeated snare resections and laser treatments so that the patient had minimal or no symptoms, the treatment was classified as partially successful. If the adenoma progressed, the treatment was classified as unsuccessful. Data were stored and analysed with SPSS (Statistical Product and Service Solutions, Chicago, Illinois, USA) version 5.0. Differences in proportions were tested with chisquare test and p-values less than or equal to 0.05 were considered significant. The study was conducted in accordance with the principles of the Helsinki Declaration and approved by the regional committee for medical research ethics in Norway. Results In the course of the 0-year period a total of 70 patients with rectal tumours treated with laser ablation were registered. Of these 7 had undergone palliative debulking of advanced inoperable rectal cancers, and were excluded from the study. Ninety-nine patients underwent snare resection and laser ablation with curative intent for lesions considered to be benign. Seven of these dropped out of follow-up examinations. The study group therefore consisted of 92 patients, 54 men (59%) and 38 women, with a median age of 72 (range 4 88) years. Thirteen (4%) patients had a recurrent tumour after previous transanal excision. The remaining 79

3 Figure Figure 2 Endoscopic snare resection followed by laser ablation in the treatment of large, sessile rectal adenomas No of patients No of snare resections 6 No of patients No of laser treaments 6 7 Fig.. Number of snare resections. Fig. 2. Number of laser treatments. patients had had no previous treatment for the polyp. All polyps were villous or tubulovillous, and the Figure polyp 2 diameter was median 4 (range 2 9) cm (Table ). The polyps were located 2 5 cm above the anal verge (Table 2). A median of two (range 6) snare resections (Fig. ) and a median of one (range 7) laser treatments (Fig. 2) were performed for each patient. The histopathological findings are summarized in Table Complete removal of the polyp was achieved in 86 (93%) patients (Fig. 3), and these were enrolled in the follow-up program. In six patients the lesion could not be completely removed (initial failure). Two of 40 these underwent rectal resection (one low anterior resection and one TEM). In four patients with incomplete removal, further curative surgery was consid- 30 ered inadvisable due to the patient s high age and 20 frailty, and they were given repeated laser treatments 8 with the aim of keeping the lesion small and minimize the symptoms. 2 This goal was achieved in two 0 of the patients (partial success), 4 but 3 in the other two 0 the neoplasm progressed (Fig. 3) During a median follow-up of 26 months (range No of laser treaments 5 80), 20 of the 86 (23%) patients with initial complete No of patients Table Polyp size and final treatment result (n = 88 with known size). Diameter No Failure Partial No Unknown success recur- final rence status < 2.5 cm 09 (0) (80) cm 54 (6)0 5 (9)0 45 (85) 3 > 5 cm 25 (29)0 4 (6) 3 7 (68) Total 88 (00) 9 (0) 5 (6) 70 (80) 4 (4) Figures in parenthesis is per cent Table 2 Polyp location (cm above dentate line). Location No (%) 0 4 cm 34 (37) 5 7 cm 7 (9) (26) > cm 7 (8) Table 3 Histopathological findings. WHO classification TNM vienna No % classification (25) Low- / moderate grade dysplasia (Low-grade intraepithelial neoplasia) pt Severe / high-grade dysplasia (High-grade intraepithelial neoplasia) ptis Intramucosal neoplasia ptis Possible submucosal infiltration pt 4.3

4 Figure 3 02 A. Nesbakken, J. Kristinsson, A. Svindland, O. C. Lunde All patients studied 92 (00%) Initial success 86 (93%) Initial failure 6 (7%) Unknown course 4 (4%) Recurrence 20 (22%) Repeated laser 4 (4%) Further treatment 2 (2%) Rectal resection 4 (4%) Repeated laser 2(3%) Progression (%) Partial success 3 (3%) No new recurrence 8 (9%) No recurrence 66 (72%) Partial success 2 (2%) Progression 2 (2%) Fig. 3. Flow chart of treatment and results. All figures in parentheses are percentages of the total (n = 92). removal of the adenoma experienced a recurrence, which was diagnosed a median of 8 (range 5 57) months after the primary treatment. Five-year estimated recurrence rate according to the Kaplan-Meier calculation was 40%. Polyp size was a risk factor for recurrence, but not the grade of dysplasia. Polyps less than 2.5 cm in diameter had a recurrence rate of zero, polyps with cm diameter had a recurrence rate of 2%, and polyps larger than 5 cm in diameter had a recurrence rate of 4% (p = 0.02). In eight of the 20 patients treated for recurrence no further recurrence was observed after the second laser ablation. Thus on an intention-to-treat basis snare resection and laser ablation could be considered successful in 74 of 92 patients (80%). In four of the 20 patients treated for recurrence, eradication was not achieved after repeated laser treatments. In three of these, however, the lesion was successfully kept small, thus minimizing the symptoms (partial success). Altogether, laser treatment could therefore be considered partially successful in five patients (5%) (Fig. 3). The relation between polyp size and final result is presented in Table. Four patients with recurrence underwent rectal resection, two of whom were found to have adenocarcinoma infiltrating all the way through the bowel wall (stage pt3). The disease course after repeated laser treatment for recurrence was not known in four patients. Five patients (5%) had complications. Four had postoperative bleeding and were readmitted to hospital. The bleeding stopped spontaneously and reoperation was not necessary. One patient developed a rectovaginal fistula. Discussion The main finding in the present study was that a high proportion of patients with large rectal adenomas can be successfully treated by the relatively simple method of snare resection and laser treatment. Conventional snare resection alone carries a high risk of recurrence, generally around 40%. Two different methods of laser treatment have been described. Some endoscopists use laser vaporization without preparatory snare resection. Most of the tissue is then destroyed without histological examination having been performed, and therefore the invasiveness cannot be assessed. Furthermore, laser treatment of large bulky masses is difficult and time-consuming. In our view, which is supported by most reports, preparatory snare resection should always be performed. Most of the polyp can usually be resected piecemeal, even in patients with very large polyps. Larger pieces can be resected if the intraluminal air is sucked out before closing the snare. Submucosal instillation of saline may also be of help.

5 Endoscopic snare resection followed by laser ablation in the treatment of large, sessile rectal adenomas 03 The outcome in the present study, with an initial failure rate of 7% and a recurrence rate following initial eradication of 23%, is fairly good, especially when considering that half of the recurrences were eradicated after new laser treatment. The results are comparable to other series, which report initial success in 6 94% of the patients, and subsequent recurrence in 9 37% of the patients (6;9 7). Like others, we found that polyp size was of prognostic significance (6;3 5;8), whereas grade of dysplasia was not, as found by Navaud (9), but contrary to what was found by Brunetaud (0) It is not possible to compare these results with those reported using other methods due to different case-mix, lack of randomized trials, small patient numbers and wide variation in reported results. Local recurrence rates of 3 4% have been reported after EMR (20;2). Many patients who present with a large rectal adenoma are old and frail, and should preferably be treated by simple methods with a minimal risk of complications. Snare resection followed by laser ablation is therefore an appropriate treatment in such cases. Forty per cent of our patients were over 80 years old, and general anaesthesia was never used. There were few adverse effects and no life-threatening complications in this series. Some of our old patients with initial failure or recurrence were so frail that treatment with curative intention could not be given, and the goal in those patients was palliation. In five of the eight patients who had repeated laser treatment with the aim of minimizing symptoms the treatment could be considered a partial success; the adenoma was kept small and the patients had minimal symptoms such as bleeding, excessive mucous discharge, diarrhoea or incontinence. Snare resection followed by laser ablation is also attractive from an economical point of view. Each session rarely takes more than 5 minutes and can be performed as an out-patient procedure. Although the initial expenditure is high due to the high cost of laser equipment, laser treatment can be used for a number of diseases. Today the most widely used alternative ablation method is probably argon plasma coagulation (APC), which is comparable to laser in terms of the strain on the patient and the simplicity of the method. Nd:YAG laser has higher penetration than APC and is more effective at a greater depth; on the other hand it also carries a higher risk of perforation. It is probably easier to estimate the depth of penetration with APC, which is a great advantage in the colon and the intra-abdominal part of the upper rectum, where bowel perforation can lead to intra-abdominal sepsis if it is not discovered immediately. It is hard to find large studies with adequate follow-up to estimate recurrence rate after endoscopic resection and APC. Zlatani et al (22) found a recurrence rate of some 50% in 47 patients with piecemeal resection alone, and similar recurrence rate in 30 patients initially treated by piecemeal resection followed by APC. Regula et al (2), in a group of 63 patients, obtained complete initial eradication after piecemeal resection and APC in 90%, and a recurrence rate of 4% at follow-up. At our hospital snare resection and laser has now largely been replaced by TEM, which has significant advantages over laser treatment. It enables a fullthickness rectal resection to be performed, and yields a perfect excision biopsy for accurate histopathological evaluation. The recurrence rate is lower in the case of excision of large adenomas; most studies report around 5%(23). Even in some patients with T cancer full-thickness TEM excision with wide margins may be curative(24). However, TEM has some disadvantages compared with laser treatment. The procedure is more resource-demanding and is usually performed under general anaesthesia, and therefore carries a higher risk for the patient. The main concern when using snare resection and laser ablation in fit patients is the risk of overlooking an early cancer. Small foci of infiltrating carcinoma can be missed preoperatively, both in forceps biopsies and in piecemeal snare resection specimens. If infiltrating carcinoma is found preoperatively, colonoscopy-based therapy is in our view not warranted if the treatment intention is curative. When infiltration has not been demonstrated, we consider that snare resection and laser ablation is a valuable treatment option in patients who are old and frail. The risk of recurrence must be kept in mind, and close follow-up with endoscopy, digital examination and biopsies is essential after laser treatment. Even then, however, there is a risk of missing an infiltrating carcinoma, especially since a recurrent carcinoma does not always present with visible mucosal changes and may develop within or outside the bowel wall. This was illustrated by two of our patients who had recurrence with infiltrating carcinoma. In both patients the diagnosis of recurrence was not made until the cancer had become advanced. In conclusion, snare resection followed by laser ablation is a simple procedure for an experienced endoscopist and well tolerated by the patient. In our view it still has a place in the treatment of frail patients with large, flat rectal neoplasms without signs of infiltrating carcinoma, and a success rate of 70 80% can be expected. However, due to the risk of missing an infiltrating carcinoma at the preoperative examination and in the resected specimen, a higher recurrence rate than TEM, and the risk of late detection of recurrent carcinoma during follow-up, this procedure is not advisable for treating large adenomas in fit patients. For these patients more radical procedures, like ESD or transanal endoscopic microsurgery, securing removal of all neoplastic tissue, are preferable. References. Boix J, Lorenzo-Zúñiga V, Moreno de Vega V et al: Endoscopic removal of large sessile colorectal adenomas: is it safe and effective? Dig Dis Sci 2007;52: Regula J, Wronska E, Polkowski M et al: Argon plasma coagulation after piecemeal polypectomy of sessile colorectal adenomas: long-term follow-up study. Endoscopy 2003; Repici A, Tricerri R: Endoscopic polypectomy: techniques, complications and follow-up. Tech Coloproctol 2004; 8 Suppl 2:

6 04 A. Nesbakken, J. Kristinsson, A. Svindland, O. C. Lunde 4. Kudo S, Tamegai Y, Yamano H et al: Endoscopic mucosal resection of the colon: the Japanese technique. Gastrointest Endosc Clin N Am 200;: Probst A, Golger D, Arnholdt H et al: Endoscopic submucosal dissection of early cancers, flat adenomas, and submucosal tumors in the gastrointestinal tract. Clin Gastroenterol Hepatol 2009;7: Brunetaud JM, Maunoury V, Cochelard D et al: Endoscopic laser treatment for rectosigmoid villous adenoma: factors affecting the results. Gastroenterology 989;97: Pigot F, Bouchard D, Mortaji M et al: Local excision of large rectal villous adenomas: long-term results. Dis Colon Rectum 2003;46: Middleton PF, Sutherland LM, Maddern GJ: Transanal endoscopic microsurgery: a systematic review. Dis Colon Rectum 2005;48: Aubert A, Meduri B, Fritsch J et al: Endoscopic treatment by snare electrocoagulation prior to Nd:YAG laser photocoagulation in 85 voluminous colorectal villous adenomas. Dis Colon Rectum 99;34: Brunetaud JM, Maunoury V, Cochelard D et al: Endoscopic laser treatment for rectosigmoid villous adenoma: factors effecting the results. Verh Dtsch Ges Inn Med 99;97:6 68. Hintze RE, Adler A, Veltzke W: Endoscopic resection of large colorectal adenomas: a combination of snare and laser ablation. Endoscopy 995;27: Low DE, Kozarek RA: Snare cautery debridement prior to Nd:YAG photoablation improves treatment efficiency of broadbased adenomas of the colorectum. Gastrointest Endosc 989; 35: Mathus-Vliegen EM, Tytgat GN: The potential and limitations of laser photoablation of colorectal adenomas. Gastrointest Endosc 99;37: Ruget O, Burtin P, Ben Bouali AK et al: Laser therapy of villous rectocolonic tumors. Evaluation of results by multivariate analysis. Gastroenterol Clin Biol 993;7: Spinelli P, Dal Fante M, Mancini A: Current role of laser and photodynamic therapy in gastrointestinal tumors and analysis of a 0-year experience. Semin Surg Oncol 992;8: Spinelli P, Mancini A, Dal Fante M: Endoscopic treatment of gastrointestinal tumors: indications and results of laser photocoagulation and photodynamic therapy. Semin Surg Oncol 995;: Norberto L, Polese L, Angriman I et al: Laser photoablation of colorectal adenomas: a 2-year experience. Surg Endosc 2005; 9: Aubert A, Meduri B, Fritsch J et al: Colorectal villous tumors. Treatment by electro-resection and laser photocoagulation. 34 cases. Presse Med 992;2: Naveau S, Perrier C, Zourabichvili O et al: Nd:YAG laser treatment of colorectal villous tumors. Gastroenterol Clin Biol 988; 2: Hurlstone DP, Sanders DS, Cross SS et al: Colonoscopic resection of lateral spreading tumours: a prospective analysis of endoscopic mucosal resection. Gut 2004;53: Fukami N, Lee JH: Endoscopic treatment of large sessile and flat colorectal lesions. Curr Opin Gastroenterol 2006;22: Zlatanic J, Waye JD, Kim PS et al: Large sessile colonic adenomas: use of argon plasma coagulator to supplement piecemeal snare polypectomy. Gastrointest Endosc 999;49: Ramirez JM, Aguilella V, Gracia JA et al: Local full-thickness excision as first line treatment for sessile rectal adenomas: long-term results. Ann Surg 2009;249: Baatrup G, Endreseth BH, Isaksen V et al: Preoperative staging and treatment options in T rectal adenocarcinoma. Acta Oncol 2009;48: Schlemper RJ, Riddell RH, Kato Y et al: The Vienna classification of gastrointestinal epithelial neoplasia. Gut 2000;47: Received: May 0, 200 Accepted: November 23, 20